Implications of Provider Network Design for Access, Affordability and Competition in Health Insurance Abstract Insurers have increasingly adopted restricted provider network insurance designs in response to rising health costs and to regulatory mandates on the scope and generosity of private insurance plans. Restricted network plans aim to lower premiums and other plan costs by limiting in-network services to a narrow set of (often low- cost) providers, or through the adoption of tiered networks that encourage use of preferred (lower-cost) facilities and providers. These plan designs have proliferated in recent years: over half of 2015 Affordable Care Act marketplace plans had narrow hospital networks (i.e., fewer than 30% of local hospitals), and one-third of large employer plans had at least one alternative tiered and/or narrow network offering in 2015. While these plan designs have been shown to lower costs, they have also raised concerns over patient access to care and whether such plans compromise affordability of care for patients who wish to maintain their current providers when they switch insurance plans. Research on insurers' provider networks has been quite limited, however, due to well-known inaccuracies and delays in provider network reporting on individual insurers' websites or in plan brochures. This project will overcome these limitations by drawing on novel cleaned and harmonized data on provider networks across a variety of markets (nongroup, employer, Medicare and Medicaid) in all 50 states. We will link these data with a variety of other data sources to execute a project that examines (Aim 1) the relationship between provider network breadth and measures of insurer and provider market concentration and consolidation; (Aim 2) the connectivity of provider networks across insurers and providers within market segments, and across market segments; (Aim 3) the quality and characteristics of in-network providers in narrow and tiered network plans; and (Aim 4) how plan networks adapt to local market changes (i.e., whether plans strategically reshape their provider networks to avoid high-cost patients when a competing plan leaves the market). Our project will inform not only scientific understanding of the role of provider networks in insurance design, access and affordability, but also will provide key insights for policymakers and regulators interested more efficient approaches to provider network oversight.